Revision history [back]

To add more info about why you should not use Haar cascade classifiers: - obsolete method - no viewpoint robustness, try detecting faces with the OpenCV faces Haar files: slight rotation and it will not detect, slight profile face and it will not detect also, ... - long training time, especially with Haar - no ratio robustness (ratio of the bounding box) - crappy detection performance, impossible to get good detection accuracy without huge number of samples and without good experience

Deep learing: - de facto standard nowadays - lots of tools - have a look at tiny-YOLO, SqueezeNet, etc. for lightweight networks for embedded platforms

To add more info about why you should not use Haar cascade classifiers: - classifiers:

• obsolete method - method
• no viewpoint robustness, try detecting faces with the OpenCV faces Haar files: slight rotation and it will not detect, slight profile face and it will not detect also, ... - ...
• long training time, especially with Haar - Haar
• no ratio robustness (ratio of the bounding box) - box)
• crappy detection performance, impossible to get good detection accuracy without huge number of samples and without good experience

Deep learing: - de facto standard nowadays - lots of tools - have a look at tiny-YOLO, SqueezeNet, etc. for lightweight networks for embedded platforms

To add more info about why you should not use Haar cascade classifiers:

• obsolete method
• no viewpoint robustness, try detecting faces with the OpenCV faces Haar files: slight rotation and it will not detect, slight profile face and it will not detect also, ...
• long training time, especially with Haar
• no ratio robustness (ratio of the bounding box)
• crappy detection performance, impossible to get good detection accuracy without huge number of samples and without good experience

Deep learing: - learing:

• de facto standard nowadays - nowadays
• lots of tools - tools
• have a look at tiny-YOLO, SqueezeNet, etc. for lightweight networks for embedded platforms